Association between acute-phase glycaemic metrics and stroke-associated pneumonia in patients with acute ischaemic stroke

Background Despite advances in acute ischaemic stroke (AIS) care, stroke-associated pneumonia (SAP) remains a common infectious complication. In this study, we evaluated the relationship between SAP and acute-phase glycaemic variability in patients with AIS. Methods We analysed 297 consecutive patients with AIS who underwent blood glucose monitoring four times daily for a minimum of 5 days. SAP was defined based on the modified Centers for Disease Control and Prevention criteria. The mean of all glucose measurements was ‘BG_mean’. Glycaemic variability was assessed using several metrics, including the SD of glucose values (BG_SD), coefficient of variation (BG_CV), time in range, hyperglycaemia ratio and mean amplitude of glycaemic excursion (MAGE). Results SAP occurred in 51 individuals (17.2%) of the 297 patients. In the multivariable logistic regression analysis, BG_mean (adjusted OR (aOR)=1.47, 95% CI 1.17 to 1.84) remained a significant predictor of SAP after adjusting for confounders. BG_SD (aOR=1.54, 95% CI 1.10 to 2.15), hyperglycaemia ratio (aOR=1.03, 95% CI 1.01 to 1.05) and MAGE (aOR=1.24, 95% CI 1.04 to 1.49) were strongly associated with SAP, independently of BG_mean. To improve the predictive power of the original A2DS2 score for SAP, we developed new predictive models by assigning additional points based on the cut-off values of individual glycaemic metrics. Among these models, only the model incorporating MAGE showed a significant improvement compared with the original A2DS2 score (p=0.03). Conclusions Elevated acute-phase glucose levels and glycaemic variability are associated with SAP in patients with AIS.